Apra Fleet

•
108
•

Orchestrates AI coding agents across networked machines, enabling their coordination from a single conversational interface.

Apra Fleet is an open-source Model Context Protocol (MCP) server designed for orchestrating large language model (LLM) agents and automating agentic workflows. It empowers multi-agent systems to coordinate tasks across various machines, whether local processes or remote machines via SSH, all managed from a single conversation. Built for developers using advanced AI coding assistants, it eliminates the need for complex dashboards or YAML configurations, allowing users to simply tell their agent what to do, and Apra Fleet handles the underlying infrastructure coordination, including agent memory persistence, secure remote execution, and dynamic cloud compute management.

Características Principales

01Coordinate AI coding agents across multiple local or remote machines via SSH.
02Securely provision scoped Git authentication and other credentials to fleet members.
03Automate cloud compute instance management with demand-based start/stop for cost efficiency.
04Enable parallel execution of tasks by multiple agents across different machines.
05Utilize an optional Project Manager skill for orchestrating multi-step, complex workflows.
0626 GitHub stars

Casos de Uso

01Run test suites on a Linux server while developing on a macOS machine.
02Orchestrate multiple agents to concurrently build frontend, backend, and run tests for an application.
03Leverage powerful cloud VMs for computationally intensive tasks like compilation while coding from a local laptop.